TeleRAG: Efficient Retrieval-Augmented Generation Inference with Lookahead Retrieval

Chien-Yu Lin, Keisuke Kamahori, Yiyu Liu, Xiaoxiang Shi, Madhav Kashyap, Yile Gu, Rulin Shao, Zihao Ye, Kan Zhu, Rohan Kadekodi, Stephanie Wang, Arvind Krishnamurthy, Luis Ceze, Baris Kasikci

Keywords: RAG, Prefetching, Acceleration, IVF Index

TL;DR: TeleRAG enhances RAG inference performance by predicting and prefetching retrieval data from CPU to GPU during LLM generation, efficiently scaling even in memory-constrained and multi-GPU environments.

Abstract:

Retrieval-augmented generation (RAG) extends large language models (LLMs) with external data sources to enhance factual correctness and domain coverage.
Modern RAG pipelines rely on large datastores, creating a significant system challenge: achieving high throughput and low latency is difficult, especially when GPU memory is limited.
To address these challenges, we propose TeleRAG, an efficient inference system that reduces latency and improves throughput with minimal GPU memory requirements.
The core innovation of TeleRAG is *lookahead retrieval*, a prefetching mechanism that predicts required data and transfers them from CPU to GPU in parallel with LLM generation.
In addition, TeleRAG adopts a prefetching scheduler and a cache-aware scheduler to support efficient multi-GPU inference with minimal overhead.
Evaluations show TeleRAG achieves up to a 1.98$\times$ average end-to-end latency reduction (single-query) and 1.83$\times$ higher average throughput (batched), as well as good scalability in throughput. This confirms the practical utility of TeleRAG for faster and more memory-efficient deployments of RAG applications.

Topics: Agentic Systems: Systems optimizations for agentic AI applications, Algorithms: Efficient algorithms for serving LLMs and generative models, Model Serving: System optimizations for model serving

Submission Number: 42